Kitzmann P et al. (2017), A key role for foxQ2 in anterior head and centr...

ECB-ART-45697

Development 2017 Aug 15;14416:2969-2981. doi: 10.1242/dev.147637.

Show Gene links Show Anatomy links

A key role for foxQ2 in anterior head and central brain patterning in insects.

Kitzmann P , Weißkopf M , Schacht MI , Bucher G .

???displayArticle.abstract???
Anterior patterning of animals is based on a set of highly conserved transcription factors but the interactions within the protostome anterior gene regulatory network (aGRN) remain enigmatic. Here, we identify the red flour beetle Tribolium castaneum ortholog of foxQ2 (Tc-foxQ2) as a novel upstream component of the aGRN. It is required for the development of the labrum and higher order brain structures, namely the central complex and the mushroom bodies. We reveal Tc-foxQ2 interactions by RNAi and heat shock-mediated misexpression. Surprisingly, Tc-foxQ2 and Tc-six3 mutually activate each other, forming a novel regulatory module at the top of the aGRN. Comparisons of our results with those of sea urchins and cnidarians suggest that foxQ2 has acquired more upstream functions in the aGRN during protostome evolution. Our findings expand the knowledge on foxQ2 gene function to include essential roles in epidermal development and central brain patterning.

???displayArticle.pubmedLink??? 28811313
???displayArticle.pmcLink??? PMC5592812
???displayArticle.link??? Development

Species referenced: Echinodermata
Genes referenced: foxe3l LOC100887844 LOC100893907 LOC105438433 LOC115919910 LOC581235 LOC583082 LOC586604 LOC594353 pole six4 six6 stk36

???attribute.lit??? ???displayArticles.show???

	Fig. 1. Tc-foxQ2 RNAi phenotype and off-target control. (A,B) Knockdown of Tc-foxQ2 with the two non-overlapping dsRNA fragments Tc-foxQ2RNAi_a (A) and Tc-foxQ2RNAi_b (B) leads to comparable proportions of cuticle phenotypes. (C,D) Detailed analysis of the head defects shows that the Tc-foxQ2RNAi_a dsRNA fragment leads to a qualitatively comparable but quantitatively stronger phenotype, marked by more intermediate and strong head defects. (E-H) L1 cuticle heads representing the different classes of head defect. Dorsal view, anterior left. (E) WT cuticle with the labrum marked in cyan, two labrum setae (yellow dots), two clypeus setae (orange dots) and two anterior vertex setae (red dots). (F) Weak head defect with a reduced labrum and at least one deleted labrum seta. (G) Intermediate head defect, additionally lacking at least one of the anterior vertex seta. (H) Strong head defect with a strongly reduced labrum, one labrum seta and one clypeus seta. In the strongest phenotypes the labrum and the anterior vertex setae are deleted. Non-specific defects refers to local alterations in cuticular structures that are most likely not due to the RNAi treatment. Strong defects refers to severe alterations affecting all tagma that are most likely not due to the RNAi treatment.
	Fig. 2. Cell death in the Tc-foxQ2RNAi phenotype. (A) Morphology of WT (Aa,a′,c,c′) and Tc-foxQ2RNAi (Ab,b′,d,d′) embryos is visualized by nuclear staining (DAPI, gray). Anterior is left in 10× and up in 40× panels. Aa,b represent late elongating germ bands; Ac,d show early retracting germ bands. The labrum is marked in blue. Tc-foxQ2RNAi embryos (6-26 h AEL) show decreased labral buds, which appear to fuse prematurely (Ab′,d′). (B) For quantification of cell death, a region of interest (region 1) and a control region (region 3) were defined. Apoptotic cells were monitored by Dcp-1 antibody staining (green). A fully elongated Tc-foxQ2RNAi WT germ band with most apoptotic cells within the labral region (Ba, region 1) is compared with a Tc-foxQ2RNAi embryo that shows most apoptotic cells within region 1 (Bb, arrowhead). (C) Box plot depicting the normalized number of apoptotic cells at five different embryonic stages for untreated (WT) and Tc-foxQ2RNAi embryos (RNAi). The values are normalized by region 3 values. Germ rudiments (stage 1) to intermediate elongating germ bands (stage 3), as well as early retracting germ bands (stage 5), show no significant increase in the number of apoptotic cells. Stage 1, P=0.33 (WT n=3, RNAi n=7); stage 2, P=0.63 (WT n=11, RNAi n=12); stage 3, P=0.19 (WT n=9, RNAi n=19); stage 5, P=0.15 (WT n=12, RNAi n=11). However, fully elongated germ bands showed significantly more apoptotic cells (***P=4.1×10−4) in Tc-foxQ2RNAi embryos (n=15) than in untreated embryos (n=17). The line inside the box represents the median, the box is defined by the first and the third quartiles, the whiskers are defined as within 1.5× the interquartile range of the lower/upper quartile and dots represent outliers. ns., not significant.
	Fig. 3. Loss of Tc-foxQ2 function leads to brain defects. (A-C) Glial tissue (black) is visualized in L1 larvae by the transgenic brainy reporter line in WT (A) and Tc-foxQ2RNAi (B,C). Voltex projections. (D-H) Mushroom bodies (MBs; black) are visualized by the transgenic MB-green reporter line in WT (D,D′) and Tc-foxQ2RNAi (E-H). MAX projections; D′, 3D projection. (A) WT L1 larval brain showing the two brain hemispheres with MBs (magenta), antennal lobes (cyan), and the midline-spanning central body (CB; yellow). (B) A weak Tc-foxQ2RNAi larval brain phenotype showing loss of the boundary between the medial lobes of the MB (compare open arrowheads in B and A). The CB appears to be reduced in size. (C) Intermediate phenotype Tc-foxQ2RNAi larval brains appear to lack the MBs and the CB appears reduced in size. (B,C) In both phenotypes the brain hemispheres appear to be fused (solid arrowhead). (D) WT L1 larval MBs in dorsal view. (D′) 3D projection of WT L1 larval MBs, providing an overview of the organization of the structures. (E) A weak Tc-foxQ2 RNAi-induced MB phenotype, where the two medial lobes appear to be fused (compare arrowheads in D and E). (F) MBs with distorted pedunculi, leading to a loss of contact between the two medial lobes (arrowheads), and reduced vertical lobes (arrow). (G) Phenotype with interdigitating MBs. (H) In strong phenotypes the MB structures are highly reduced or absent. OL, optical lobe; mL, medial lobe; Pe, pedunculus; vL, vertical lobe; Ca, calyx.
	Fig. 4. Tc-foxQ2 is expressed in a highly dynamic pattern at the anterior pole. Expression of Tc-foxQ2 in WT embryos monitored by whole-mount in situ hybridization (ISH). Anterior is up in all images. (A) Tc-foxQ2 expression starts with the formation of the germ rudiment. (A-E) The early Tc-foxQ2 expression starts with two domains located at the anterior pole, which successively approach each other, probably as a consequence of morphogenetic movements. (F) The expression splits into several domains in late elongating germ bands, with expression in the putative neuroectoderm (arrow, presumably including parts of the pars intercerebralis; Posnien et al., 2011b) and in the labral/stomodeal region (arrowhead). (G) The expression domains flanking the prospective stomodeum become more prominent (arrow). (H) The anterior median expression domain frames the lateral parts of the labral buds (arrowhead). (I,J) At fully elongated and early retracting germ band stages the two expression domains flanking the stomodeum become posteriorly linked (arrow). (K) Staining with the more sensitive TSA-Dylight550 reveals four dot-like expression domains in the ocular region (arrowheads). (L) At retracting germ band stages Tc-foxQ2 is expressed in a narrow U-shaped pattern and the neuroectodermal expression domains are reduced in size (arrow).
	Fig. 5. Co-expression of Tc-foxQ2 and anterior head patterning genes. Expression is visualized by double ISH, using NBT/BCIP (blue) and TSA-Dylight550 (red). Regions of co-expression are outlined. Anterior is up in all images. (Aa-g) Co-expression of Tc-foxQ2 with Tc-wg is found only after full elongation (Af,g). (Ba-g) Tc-foxQ2 and Tc-six3 expression completely overlap during germ rudiment stages (Ba). In early elongating germ bands the co-expression is limited to a narrow lateral stripe of the anterior median region (AMR) (Bb). Intermediate germ bands show mutually exclusive expression of Tc-foxQ2 and Tc-six3 (Bc). At later stages, expression overlap is found within the neurogenic region (Bd,e) and later also in the labral buds (Bf,g). (Ca-g) Tc-cnc expression is completely overlapping with Tc-foxQ2 at early embryonic stages (Ca-c) but later co-expression is found in the labral/stomodeal region (Cd-g). (Da-g) Tc-scro is partially co-expressed at early embryonic stages (Da-c). In late elongating germ bands the co-expression is restricted to a narrow lateral stripe (Dd) and the posterior portion of the labral buds, close to the stomodeum and small areas of the neurogenic region (De-g). (Ea-g) Tc-croc expression is partially overlapping with Tc-foxQ2 at early stages (Ea-c) and later in a region close to the stomodeum (Ed-f).
	Fig. 6. Mutual activation of Tc-foxQ2 and Tc-six3 and their repression by Wnt signaling. Expression pattern of Tc-foxQ2 in WT (Aa-c), Tc-six3RNAi (Ca-c), Tc-arrRNAi (Ea-c) and Tc-axinRNAi (Gb,c) embryos and expression pattern of Tc-six3 in WT (Ba-c), Tc-foxQ2RNAi (Da-c), Tc-arrRNAi (Fa-c), and Tc-axinRNAi (Ha,b) embryos monitored by ISH. (Ca-c) Tc-foxQ2 expression was lost in Tc-six3RNAi embryos. (Da) Tc-six3 expression was strongly reduced in Tc-foxQ2RNAi germ rudiments (arrowhead). (Db,c) Later, median expression emerged but the lateral Tc-six3 expression remained reduced or absent (arrowhead), whereas the ocular domain appeared unchanged. (Ea,b) Tc-foxQ2 expression was unchanged in early Tc-arrRNAi embryos but the median Tc-foxQ2 expression was reduced at later stages (Ec, solid arrowhead), whereas the neurogenic Tc-foxQ2 expression domain expanded (Ec, open arrowhead). (Fa-c) In Tc-arrRNAi embryos the Tc-six3 expression expanded towards the posterior in the neuroectoderm (arrowheads), whereas the stomodeal domain appeared to be unaffected. In Tc-axinRNAi embryos Wnt signaling is derepressed, which led to repression of Tc-foxQ2 expression (Ga,b) and to a strong reduction of Tc-six3 expression (Ha,b, arrowheads).
	Fig. 7. Tc-foxQ2RNAi regulates Tc-cnc, Tc-croc and Tc-scro expression. Expression pattern of Tc-cnc in WT (Aa-d) and Tc-foxQ2RNAi (Ba-d) embryos, expression of Tc-croc in WT (Ca-d) and Tc-foxQ2RNAi (Da-d) embryos, and expression of Tc-scro in WT (Ea-d) and Tc-foxQ2RNAi (Fa-d) embryos monitored by ISH. (Ba,b) In Tc-foxQ2RNAi embryos the anterior domain of Tc-cnc expression was reduced (arrowheads). Prior to this stage, no changes in the expression pattern were observed (not shown). (Bc,d) In fully elongated and retracting germ bands the labral expression was strongly reduced (arrowheads), while stomodeal expression was slightly altered (arrow). (Da-d) Throughout development, the Tc-croc expression pattern was lacking the anterior portion of its AMR expression domain (arrowheads). (Fa) Expression of Tc-scro is reduced to a narrow stripe along the anterior fold (arrowhead). (Fb-d) Later stages show an atypical bridging between the labral/stomodeal and the neurogenic Tc-scro expression domains (arrows).
	Fig. 8. Tc-foxQ2RNAi embryos show reduced Tc-chx, Tc-six4 and Tc-rx expression. Expression patterns of Tc-chx in WT (Aa-d) and Tc-foxQ2RNAi embryos (Ba-d), of Tc-six4 in WT (Ca-d) and Tc-foxQ2RNAi embryos (Da-d) and of Tc-rx in WT (Ea-c) and Tc-foxQ2RNAi embryos (Fa-c) as monitored by ISH. (Ba) Tc-chx expression was completely absent in early elongating Tc-foxQ2RNAi germ bands (arrowhead). (Bb-d) At later stages, the labral Tc-chx expression domains were almost absent (arrowheads), while the anterior neurogenic expression domains were strongly reduced (arrows). The ocular Tc-chx expression domain remained unaffected. (Da) Expression of Tc-six4 is strongly reduced in early elongating germ bands (arrow). (Db-d) At later stages, only the median posterior extensions of the Tc-six4 expression domains were reduced (arrows). (Fa) Tc-rx expression is strongly reduced or completely absent in early elongating Tc-foxQ2RNAi germ bands. (Fb,c) At later stages, the neurogenic Tc-rx expression pattern appeared unaffected, but the labral expression domains were absent (Fb, arrowhead) or reduced in size (Fc, arrowhead). Anterior is up in all images.
	Fig. 9. Ectopic Tc-foxQ2 expression impacts head patterning. Expression of head patterning genes in heat shock-treated WT (A,D,D′,G,J,M) and hsp68-Tc-foxQ2 (B,C,E,E′,F,F′,H,I,K,L,N,O) embryos (14-18 h AEL) as monitored by ISH. (B,C) Ectopic Tc-foxQ2 expression led to reduced Tc-rx expression. (E,F) Tc-six4 expression showed a premature onset at the anterior tip (arrows; compare embryonic stages of E′,F′ with D′). Further, this premature expression was expanded; and an additional, more posterior Tc-six4 expression domain emerged (arrowheads). (H,I) Tc-scro expression started prematurely and was expanded. (K,L) By contrast, early elongating germ bands showed reduced Tc-scro expression – presumably a secondary effect. (N,O) The anterior Tc-cnc expression domain spread towards the posterior (arrowheads). The mandibular Tc-cnc expression domain was reduced and became somewhat punctate (arrows). Anterior is up in all panels except D′-F′, where anterior is to the left.
	Fig. 10. Tc-foxQ2 and Tc-six3 form a regulatory module in the aGRN. Black lines indicate previously reported interactions (based on: Kittelmann et al., 2013; Posnien et al., 2011b; Schaeper et al., 2010; Siemanowski et al., 2015). Arrows represent gene activation, and cross-bars indicate gene repression. This aGRN represents the interactions at early embryonic stages – later interactions are likely to differ. (1) Tc-six3 is the most upstream factor for patterning the anterior median head and neuroectoderm. (2) Tc-foxQ2, like Tc-six3, is a key player in anterior head development, with a somewhat later onset of expression than Tc-six3. Mutual activation, similar effects on other genes and similar phenotypes suggest that they form a regulatory module (indicated by the dotted circle). (3) Tc-foxQ2 and Tc-six3 are both antagonized by Wnt/β-catenin signaling. (4) Tc-six3 acts on Notch signaling via Tc-ser. (5/6) Mutual repression of Tc-six3 and ocular Tc-wg expression. (7) Notch signaling-dependent activation of Tc-foxQ2 is restricted to lateral parts of the anterior Tc-foxQ2 expression. (8) Regulatory activities of Tc-foxQ2 and Tc-six3 are similar with respect to several downstream targets. (9) Tc-foxQ2 but not Tc-six3 is repressed by Tc-croc activity. (10) An unknown factor ‘X’ is predicted to activate the posterior part of the Tc-croc expression, while Tc-six3 and Tc-foxQ2 are required for the anterior portion. (11) Tc-rx is repressed by Tc-foxQ2 but is not regulated by Tc-six3. Note that the assumed repressive effect is based on non-overlapping expression of these genes and the effects found in misexpression experiments (see text for conflicting RNAi data). (12) The late effect of Tc-foxQ2 on Tc-scro, observed in gain-of-function experiments, is most likely secondary and is, hence, not considered here. (13) Notch signaling is involved in labrum development by regulating cell proliferation.

References [+] :

Altschul, Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. 1997, Pubmed